Towards automated aircraft maintenance inspection. A use case of detecting aircraft dents using mask r-cnn

Towards automated aircraft maintenance inspection. A use case of detecting aircraft dents using mask r-cnn

Bouarfa, Soufiane (Abu Dhabi Polytechnic, Al Ain)
Doğru, Anıl (Özyeğin University)
Arizar, Ridwan (Singulair Solutions B.V., Rotterdam)
Aydoğan, Reyhan (TU Delft Interactive Intelligence; Özyeğin University)
Serafico, Joselito (Abu Dhabi Polytechnic, Al Ain)

2020

Deep learning can be used to automate aircraft maintenance visual inspection. This can help increase the accuracy of damage detection, reduce aircraft downtime, and help prevent inspection accidents. The objective of this paper is to demonstrate the potential of this method in supporting aircraft engineers to automatically detect aircraft dents. The novelty of the work lies in applying a recently developed neural network architecture know by Mask R-CNN, which enables the detection of objects in an image while simultaneously generating a segmentation mask for each instance. Despite the small dataset size used for training, the results are promising and demonstrate the potential of deep learning to automate aircraft maintenance inspection. The model can be trained to identify additional types of damage such as lightning strike entry and exit points, paint damage, cracks and holes, missing markings, and can therefore be a useful decision-support system for aircraft engineers.

http://resolver.tudelft.nl/uuid:36cf72d5-8059-4ddf-beb4-380bdd2f3baf

https://doi.org/10.2514/6.2020-0389

American Institute of Aeronautics and Astronautics Inc. (AIAA)

978-1-62410-595-1

AIAA Scitech 2020 Forum

AIAA Scitech 2020 Forum, 2020-01-06 → 2020-01-10, Orlando, United States

AIAA Scitech 2020 Forum, 1 PartF

Institutional Repository

conference paper

© 2020 Soufiane Bouarfa, Anıl Doğru, Ridwan Arizar, Reyhan Aydoğan, Joselito Serafico